CN102567939A - Separating method and device for visible watermark in image - Google Patents
Separating method and device for visible watermark in image Download PDFInfo
- Publication number
- CN102567939A CN102567939A CN2010106222125A CN201010622212A CN102567939A CN 102567939 A CN102567939 A CN 102567939A CN 2010106222125 A CN2010106222125 A CN 2010106222125A CN 201010622212 A CN201010622212 A CN 201010622212A CN 102567939 A CN102567939 A CN 102567939A
- Authority
- CN
- China
- Prior art keywords
- image
- watermarking images
- watermark
- background watermarking
- binaryzation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Editing Of Facsimile Originals (AREA)
- Image Processing (AREA)
Abstract
The invention provides a separating method for a visible watermark in an image. The method comprises the steps of: performing binarization operation on a digital image in which the watermark is to be separated, so as to obtain a binarized image which comprises the real content of the image and using the binarized image as a foreground mask image; obtaining a primary background watermark image with reference to the foreground mask image; and correcting the primary background watermark image, wherein the correction operation comprises at least one of edge correction, binarization operation and de-noising procession. Correspondingly, the invention provides a separating device. The separating method and the separating device disclosed by the invention can, without using any specific hardware or software, automatically extract various digital images and generate semitransparent watermarks in material of the digital images in any way, thereby laying a good foundation for subsequent automatic analysis and identification of the watermarks.
Description
Technical field
The present invention relates to the digital image processing techniques field, relate in particular to the separation method and the device of visible watermark in a kind of image.
Background technology
Digital watermark technology is as an important branch of Information Hiding Techniques; Be widely used in the protected digital object of needs (like image, video, audio frequency), discern the false from the genuine, protect copyright to reach, follow the tracks of purpose tortious through some numerical information of embedding (like author's sequence number, corporate logo, product marking, significant text) in protected digital object.From the whether visible angle of watermark, can watermark be divided into two kinds of visible watermark and invisible watermarks, wherein, visible watermark has various ways, and is common like translucent background patterns and literal.Visible watermark is usually used in following several kinds of situation: 1) need not professional equipment, only just can differentiate the true and false of watermark carrier fast with people's visual inspection watermark, translucent watermarks such as " specimen page " that for example adds in the watermark in the bank note, the books, " academic probation " etc.; 2) need the hardware device of specialty to discern the entrained specifying information of watermark; The for example label on the bill, bar code; Though can see these watermarks easily with human eye, know the specifying information that these watermarks are entrained, need with the information in scanner reading tag or the bar code; 3) need be with extracting watermark so that subsequent treatment is carried out in watermark or its carrier with the corresponding software algorithm of embed watermark, for example application number is that 200610134046.8 Chinese patent " visible watermark the removed method of block local encrypted digital image ", application number are 200310103202 one Chinese patent application " a kind of new anti-counterfeiting method for file of affixing one's seal ".
For the method for separating digital figure watermark, in the prior art, generally all be at first to generate watermark according to the information that is comprised in the digital picture (like gray-scale value), and the watermark that is generated is embedded in the digital picture according to certain rule or algorithm with software; Then, in the watermark detachment process, utilize with the rule of generation watermark or the rule or the algorithm of algorithm contrary and extract watermark, to recover the original image carrier or to carry out other subsequent treatment.Therefore, the separation of translucent watermark in the numerical imaging at present must depend on specific watermarking algorithm, does not have general watermark separation algorithm.Yet; At present; Less for watermark Study on Separation in the image of the paper document scanning back formation that contains translucent watermark; Still lack effective means, and then make follow-up analysis or utilization to watermark information in the scan image need depend on artificial or specific hardware or software, automaticity is not high.
Summary of the invention
In order to overcome the above problems, the present invention provides the separation method and the device of visible watermark in a kind of image, with the translucent watermark in the automatic extraction image.
In order to realize above purpose; The separation method of visible watermark may further comprise the steps in the image provided by the invention: treat the numerical imaging that separates watermark and carry out the binaryzation operation; The image that comprises the image real content that obtains binaryzation is as prospect mask image, and obtains preliminary background watermarking images with reference to this prospect mask image; Preliminary background watermarking images is revised, and said correction comprises at least one in edge modifications, binaryzation operation and the denoising.
Preferably, utilize maximum variance between clusters to treat the numerical imaging that separates watermark automatically and carry out the binaryzation operation.
Preferably; The said step that obtains preliminary background watermarking images with reference to prospect mask image may further comprise the steps: the color of checking each pixel on the prospect mask image one by one; If the color of pixel is a black; Color of pixel corresponding with this pixel on the then said numerical imaging is set to white, otherwise keeps color of pixel corresponding with this pixel on the said numerical imaging.
Preferably, the said step that preliminary background watermarking images is revised comprises: with reference to prospect mask image the preliminary background watermarking images that obtains is carried out the edge modifications operation, to remove edge noise; The background watermarking images of revising is carried out the binaryzation operation, to remove the part clutter noise; A disconnected part that connects the background watermarking images, and this background watermarking images carried out denoising.
Preferably; Saidly the step that the preliminary background watermarking images that obtains carries out the edge modifications operation may further comprise the steps: collect the marginal point that all black and white in the prospect mask image have a common boundary and the N field point of these marginal points with reference to prospect mask image; Wherein, The N value is relevant with image resolution ratio, and resolution is high more, and the N value is big more; In the preliminary background watermarking images that obtains, be set to white with the corresponding pixel of collecting of point.
Preferably, utilize maximum variance between clusters automatically the background watermarking images of revising to be carried out the binaryzation operation.
Preferably, utilize expansive working to connect a disconnected part of the background watermarking images of binaryzation.
Preferably, utilize and to ask the method for carrying out denoising after the connected domain again that the background watermarking images is carried out denoising.
Preferably, the paper material that contains translucent watermark is scanned, with gray level image or the coloured image that obtains to quantize.
Preferably, for colour wait to separate the image that quantizes that obtains after numerical imaging or the scanning of watermark, before the binaryzation operation, be gray level image with this image transitions.
Preferably, keep the colouring information of coloured image, and utilize these colouring informations that the background watermarking images of binaryzation is reverted to colored background watermarking images.
In order to realize above method, the present invention provides the tripping device of visible watermark in a kind of image, and this device comprises: input block is used to import the numerical imaging of waiting to separate watermark; The watermark separative element is used for separating the background watermarking images of the numerical imaging of input; Output unit; Be used to export the background watermarking images that the watermark separative element is separated, wherein, the watermark separative element comprises: prospect mask code processing unit; Be used for the numerical imaging of input is carried out the binaryzation operation, the image that comprises the image real content that obtains binaryzation is as prospect mask image; The prospect mask storage location is used to store prospect mask image; The watermark extracting unit, the prospect mask image that is used for storing with reference to the prospect mask storage location obtains preliminary background watermarking images; The watermark amending unit is used for preliminary background watermarking images is revised background watermarking images that obtain being communicated with, denoising.
Preferably, this device also can comprise: gradation conversion unit is used for converting the coloured image through the input block input into gray level image; The colouring information storage unit; Be used to store each color of pixel information through the coloured image of input block input, output unit utilizes the colouring information of storing in the colouring information storage unit that the binaryzation background watermarking images that obtains is reverted to colored background watermark recovery.
The binary image that the present invention utilizes binarization method to extract mainly to comprise the image real content is as prospect mask image; And from original image, extract preliminary background watermarking images with reference to this prospect mask image; Through the preliminary background watermarking images that extracts is carried out correction operations such as edge modifications, connection, denoising, obtain background watermarking images that be communicated with, denoising then.Through this technical scheme; Can be in the translucent watermark under the situation of specific hardware or software, automatically extracting various numerical imagings (comprising the various materials that contain translucent watermark that can generate numerical imaging by all means), thus lay a good foundation for the automatic analysis of follow-up watermark, identification.
Description of drawings
Fig. 1 is the process flow diagram of the separation method of visible watermark in the image according to an embodiment of the invention;
Fig. 2 is the gray level image that paper material forms after overscanning in the embodiment of the invention;
Fig. 3 is the prospect mask image in the embodiment of the invention;
Fig. 4 is the preliminary background watermarking images in the embodiment of the invention;
Fig. 5 is the partial enlarged drawing of image shown in Figure 4;
Fig. 6 directly carries out the background watermarking images that the binaryzation operation obtains to image shown in Figure 4;
Fig. 7 is the background watermarking images of image shown in Figure 4 through obtaining after the edge modifications;
Fig. 8 is the background watermarking images of image shown in Figure 7 through obtaining after the binaryzation operation;
Fig. 9 is the background watermarking images that obtains after image shown in Figure 8 connects through disconnected pen;
Figure 10 is the background watermarking images of image shown in Figure 9 through obtaining after the denoising operation;
Figure 11 is the block diagram of the tripping device of visible watermark in the image according to an embodiment of the invention.
Embodiment
Watermark separation method according to the present invention is applicable to any numerical imaging and any material that contains translucent watermark that can generate numerical imaging through any approach that contains translucent watermark; Such as; For the paper material that contains translucent watermark; Can at first it be scanned into gray level image or the coloured image that quantizes, separate watermarking images according to the inventive method then through scanner.Below, will describe the present invention with reference to accompanying drawing and embodiment.
Fig. 1 is the process flow diagram of the separation method of visible watermark in the image according to an embodiment of the invention.In the present embodiment, so that the paper material that contains translucent watermark is handled the inventive method is described.With reference to Fig. 1, this method may further comprise the steps:
Step S1, the paper material that contains translucent watermark is scanned, obtain the gray level image that quantizes.
In this step, in order to keep the translucent watermark on the paper material, when using scanner to scan, can not scan type be set to the black and white binary map, otherwise watermark will be erased, can only scan type be set to gray level image or coloured image.If there is not multicolour pattern on the paper material, then preferred scanning is for gray level image, and is so not only faster, and the disk space that takies is littler.
The Micro biological Tests report that in the present embodiment, will contain translucent watermark is scanned into gray level image as shown in Figure 2.
Step S2, treat the numerical imaging that separates watermark and carry out the binaryzation operation, the image that comprises the image real content that obtains binaryzation is as prospect mask image.
Generally speaking, the color of translucent watermark is all more shallow, therefore, and in this step; Through the binaryzation operation translucent watermark is partly erased, that is, be set to background colour according to binaryzation algorithm translucent portion; That is, white, the main contents of original image are set to foreground; That is, black, thus make in the image the only real content of remaining image.In the numerical imaging that scanning obtains,, nearly all be set to black after these some process binaryzation operations, thereby form partial noise owing to when scanning, generally may produce the darker point of some colors in the blank place of paper.Therefore, add partial noise through the real content that generally all comprises image in the image that is obtained after the binaryzation operation.About the particular content of binaryzation operation, can be referring to " Visual C++ digital picture mode identification technology and engineering practice " (Zhang Honglin, People's Telecon Publishing House, ISBN:7-115-10951-6) the 441st page.
In the present embodiment, utilize Ostu algorithm (maximum variance between clusters) to obtain threshold value automatically, and the pixel that is lower than threshold value is set to background colour, that is, white, the pixel that is not less than threshold value is set to foreground, that is, and black.As shown in Figure 3, gray level image shown in Figure 2 has kept the body part in the former report through after the binaryzation operation, has partial noise simultaneously.Because this image plays the effect of " mask " in subsequent operation, therefore, this image is referred to as " prospect mask image ", obviously this image is identical with original image size shown in Figure 2, but only is made up of two kinds of colors of black and white.
Step S3, obtain preliminary background watermarking images with reference to prospect mask image.
In this step; Carry out prospect mask image shown in Figure 3 and original image shown in Figure 2 overlapping; Check the color of each pixel on the prospect mask image one by one, if the color of pixel is a black, color of pixel corresponding with this pixel on the original image then shown in Figure 2 is set to white; If pixel is a white, then keep color of pixel corresponding on the original image shown in Figure 2 with this pixel.This step can be seen the AND-operation of prospect mask image shown in Figure 3 and original image shown in Figure 2 as.As shown in Figure 4, in the background watermarking images that after handling through step S3, obtains, the real content on the original image is erased, and remaining part mainly is made up of translucent watermark, is called " preliminary background watermarking images ".
Below will make amendment to preliminary background watermarking images, to obtain background watermarking images that be communicated with, denoising through following step.
Step S4, preliminary background watermarking images is carried out edge modifications operation with reference to prospect mask image.
In this step; Because there is the middle transition look in pattern/literal darker on original image shown in Figure 2 with more shallow background intersection; And these points are nearly all turned to white by two-value in the prospect mask image that after handling through step S2, obtains; Therefore in step S3 with operation in, all be retained with these white point corresponding gray in the original image shown in Figure 2.Shown in the partial enlarged drawing among Fig. 5, a large amount of significantly edge noises have been produced at the text edge of original image.At this moment, then as shown in Figure 6 if these points are not made amendment but directly carry out next step binaryzation step, these generation much noises of naming a person for a particular job.
In the present embodiment; The method of revising these noise points is with reference to prospect mask figure, to collect the marginal point of all black and white boundaries in the prospect mask image shown in Figure 3 and the N field point of marginal point; N is an empirical value; Preset image resolution (Dip) is relevant during with scan image, and resolution is high more, and the N value should be provided with greatly more.Can be with reference to following piecewise function value:
In the present embodiment.Scanning resolution is 300Dpi, and then N gets 2.In preliminary background watermarking images shown in Figure 4, be set to white with the pixel of bleeding point correspondence position, then the edge transition point shown in Fig. 4 is erased basically, avoids producing a large amount of noises.Revised background watermarking images is as shown in Figure 7.
Step S5, the background watermarking images of revising is carried out the binaryzation operation, obtain the background watermarking images of denoising.
In image shown in Figure 7, pattern/literal and edge transition point thereof that color is darker are eliminated, and this image mainly is made up of watermark, but still have partial noise.In order to eliminate these clutter noise points, in this step, utilize the binaryzation operation to handle.Though these watermark point paler colours; After but process binaryzation algorithm is handled; Watermark point can be divided into black color dots basically, and partial noise and white background will be divided into white point, thereby obtains removing the background watermarking images (as shown in Figure 8) of noise.
Disconnected part in step S6, the connection background watermarking images.
Because after the background watermarking images being carried out the binaryzation operation, part more shallow in the part watermark is divided into white, thereby the partly phenomenon of disconnected pen has appearred.Therefore, need the part that occur disconnected pen in the background watermarking images shown in Figure 8 be coupled together.
In the present embodiment, utilize expansive working that the part of the disconnected pen of part is coupled together, particularly, 2 pixels for example can outwards expand.About the particular content of expansive working, can be referring to " Digital Image Processing (second edition) " (Ruan Qiuqi, Electronic Industry Press, ISBN:9787121038044) the 450th page.Background watermarking images shown in Figure 7 is as shown in Figure 9 through the image that obtains after the expansive working.
Step S7, step S6 is obtained the background watermarking images carry out the denoising operation.
After the above-mentioned steps processing, in background watermarking images shown in Figure 8, still there are some isolated, little noises, therefore, also need carry out further denoising operation.In this step; Isolated point Denoising Algorithm arbitrarily can be used, preferably, the method for asking connected domain denoising afterwards can be used; The method of asking about connected domain; For example can referring to " Digital Image Processing principle and utilization " (Tinku Acharya etc., publishing house of Tsing-Hua University, ISBN:9787302152224).The size of connected domain is set to empirical value, and in the present embodiment, preset image resolution is relevant during with scan image, and resolution is high more, and the big or small M value of connected domain should be provided with greatly more.Can be with reference to following piecewise function value:
The scanning resolution of present embodiment is 300Dpi, and then M gets 80.The background watermarking images of the revised binaryzation of gained is shown in figure 10.
Equally, as stated,, at first must be converted into gray level image, extract the translucent watermark in the gray level image of changing through step S2-S7 then for colored numerical imaging.In addition, also can keep the colouring information in the image, and utilize these colouring informations that the background watermarking images of binaryzation shown in Figure 10 is reverted to colored background watermarking images.
Below will describe the tripping device of visible watermark in the image according to the present invention with reference to Figure 11.Shown in figure 11, this device comprises: input block 10 is used to import the numerical imaging of waiting to separate watermark; Watermark separative element 20 is used for separating the background watermarking images of the numerical imaging of input; Output unit 30 is used to export the background watermarking images that the watermark separative element is separated.Wherein, watermark separative element 20 comprises: prospect mask code processing unit 21, be used for the numerical imaging of input is carried out the binaryzation operation, and the image that comprises the image real content that obtains binaryzation is as prospect mask image; Prospect mask storage location 22 is used to store prospect mask image; Watermark extracting unit 23 is used for obtaining preliminary background watermarking images with reference to the prospect mask image of prospect mask storage location 22 storages; Watermark amending unit 24 is used for preliminary background watermarking images is revised background watermarking images that obtain being communicated with, denoising.
Preferably, watermark amending unit 24 comprises: edge modifications unit 241 is used for reference to prospect mask image the preliminary background watermarking images that obtains through watermark extracting unit 23 being carried out the edge modifications operation, to remove edge noise; Binarization unit 242 is used for the background watermarking images of revising is carried out the binaryzation operation, to remove the part clutter noise; Connected unit 243 is used for connecting the disconnected part through the background watermarking images of binarization unit 242 processing; Denoising unit 244 is used for the background watermarking images of handling through connected unit 243 is carried out denoising.
Preferably, this device also can comprise: gradation conversion unit 40 is used for converting the coloured image through the input block input into gray level image; Colouring information storage unit 50 is used to store each color of pixel information through the coloured image of input block input.Output unit 30 utilizes the colouring information of storing in the colouring information storage unit that the binaryzation background watermarking images that obtains is reverted to colored background watermark recovery.
Below with reference to accompanying drawing and embodiment the present invention is described in detail; But; Should be appreciated that the present invention is not limited to above disclosed specific embodiment, modification that any those skilled in the art expects on this basis easily and modification all should be included in protection scope of the present invention.
Claims (16)
1. the separation method of visible watermark in the image may further comprise the steps:
Treat the numerical imaging that separates watermark and carry out the binaryzation operation, the image that comprises the image real content that obtains binaryzation is as prospect mask image, and obtains preliminary background watermarking images with reference to this prospect mask image;
Preliminary background watermarking images is revised, and said correction comprises at least one in edge modifications, binaryzation operation and the denoising.
2. method according to claim 1 is characterized in that, utilizes maximum variance between clusters to treat the numerical imaging that separates watermark automatically and carries out the binaryzation operation.
3. method according to claim 1 is characterized in that, the said step that obtains preliminary background watermarking images with reference to prospect mask image may further comprise the steps:
Check the color of each pixel on the prospect mask image one by one; If the color of pixel is a black; Color of pixel corresponding with this pixel on the then said numerical imaging is set to white, otherwise keeps color of pixel corresponding with this pixel on the said numerical imaging.
4. method according to claim 1 is characterized in that, the said step that preliminary background watermarking images is revised comprises:
With reference to prospect mask image the preliminary background watermarking images that obtains is carried out the edge modifications operation, to remove edge noise;
The background watermarking images of revising is carried out the binaryzation operation, to remove the part clutter noise;
A disconnected part that connects the background watermarking images, and this background watermarking images carried out denoising.
5. method according to claim 4 is characterized in that, saidly with reference to prospect mask image the step that the preliminary background watermarking images that obtains carries out the edge modifications operation may further comprise the steps:
The marginal point that all black and white in the collection prospect mask image are had a common boundary and the N field point of these marginal points, wherein, the N value is relevant with image resolution ratio, and resolution is high more, and the N value is big more;
In preliminary background watermarking images, be set to white with the corresponding pixel of collecting of point.
6. method according to claim 4 is characterized in that, utilizes maximum variance between clusters automatically the background watermarking images of revising to be carried out the binaryzation operation.
7. method according to claim 4 is characterized in that, utilizes expansive working to connect a disconnected part of the background watermarking images of binaryzation.
8. method according to claim 4 is characterized in that, utilizes and to ask the method for carrying out denoising after the connected domain again that the background watermarking images is carried out denoising.
9. method according to claim 1 is characterized in that,
Paper material to containing translucent watermark scans, with gray level image or the coloured image that obtains to quantize.
10. according to claim 1 or 9 described methods, it is characterized in that,
For colour wait to separate the image that quantizes that obtains after numerical imaging or the scanning of watermark, before the binaryzation operation, be gray level image with this image transitions.
11. method according to claim 10 is characterized in that, keeps the colouring information of coloured image, and utilizes these colouring informations that the background watermarking images of binaryzation is reverted to colored background watermarking images.
12. the tripping device of visible watermark in the image comprises:
Input block is used to import the numerical imaging of waiting to separate watermark;
The watermark separative element is used for separating the background watermarking images of the numerical imaging of input;
Output unit is used to export the background watermarking images that the watermark separative element is separated,
Wherein, the watermark separative element comprises:
Prospect mask code processing unit is used for the numerical imaging of input is carried out the binaryzation operation, and the image that comprises the image real content that obtains binaryzation is as prospect mask image;
The prospect mask storage location is used to store prospect mask image;
The watermark extracting unit, the prospect mask image that is used for storing with reference to the prospect mask storage location obtains preliminary background watermarking images;
The watermark amending unit is used for preliminary background watermarking images is revised background watermarking images that obtain being communicated with, denoising.
13. device according to claim 12 is characterized in that, said watermark amending unit comprises:
The edge modifications unit is used for reference to prospect mask image the preliminary background watermarking images that obtains through the watermark extracting unit being carried out the edge modifications operation, to remove edge noise;
Binarization unit is used for the background watermarking images of revising is carried out the binaryzation operation, to remove the part clutter noise;
Connected unit is used for connecting the disconnected part through the background watermarking images of binarization unit processing;
The denoising unit is used for the background watermarking images of handling through connected unit is carried out denoising.
14. device according to claim 12 is characterized in that, also comprises gradation conversion unit, is used for converting the coloured image through the input block input into gray level image.
15. device according to claim 12 is characterized in that, also comprises the colouring information storage unit, is used to store each color of pixel information through the coloured image of input block input.
16. device according to claim 15 is characterized in that, output unit utilizes the colouring information of storing in the colouring information storage unit that the binaryzation background watermarking images that obtains is reverted to colored background watermark recovery.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010106222125A CN102567939A (en) | 2010-12-27 | 2010-12-27 | Separating method and device for visible watermark in image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010106222125A CN102567939A (en) | 2010-12-27 | 2010-12-27 | Separating method and device for visible watermark in image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102567939A true CN102567939A (en) | 2012-07-11 |
Family
ID=46413294
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010106222125A Pending CN102567939A (en) | 2010-12-27 | 2010-12-27 | Separating method and device for visible watermark in image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102567939A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108171235A (en) * | 2018-01-08 | 2018-06-15 | 北京奇艺世纪科技有限公司 | Title area detection method and system |
| CN108550099A (en) * | 2018-04-13 | 2018-09-18 | 北京焦点新干线信息技术有限公司 | The method and apparatus of watermark in a kind of removal image |
| CN108596916A (en) * | 2018-04-16 | 2018-09-28 | 深圳市联软科技股份有限公司 | Watermark recognition methods, system, terminal and medium similar in a kind of color |
| CN109214999A (en) * | 2018-09-21 | 2019-01-15 | 传线网络科技（上海）有限公司 | A kind of removing method and device of video caption |
| CN109903210A (en) * | 2019-01-04 | 2019-06-18 | 阿里巴巴集团控股有限公司 | Minimizing technology, device and the server of watermark |
| WO2019137237A1 (en) * | 2018-01-09 | 2019-07-18 | 佛山科学技术学院 | Separate ocr recognition method and system thereof |
| CN111798359A (en) * | 2020-05-19 | 2020-10-20 | 佛山市南海区广工大数控装备协同创新研究院 | Deep learning-based image watermark removing method |
| CN111833231A (en) * | 2019-04-15 | 2020-10-27 | 阿里巴巴集团控股有限公司 | Watermark extraction method, device and system |
| CN113940090A (en) * | 2019-11-22 | 2022-01-14 | Inka 安特沃客有限公司 | Client-side evidence obtaining watermark device, system and method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101051351A (en) * | 2007-05-23 | 2007-10-10 | 重庆医科大学 | Image band parameter two-valued method and device using said method |
| CN101593276A (en) * | 2008-05-29 | 2009-12-02 | 汉王科技股份有限公司 | A kind of video OCR image-text separation method and system |
| CN101621615A (en) * | 2009-07-24 | 2010-01-06 | 南京邮电大学 | Self-adaptive background modeling and moving target detecting method |
-
2010
- 2010-12-27 CN CN2010106222125A patent/CN102567939A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101051351A (en) * | 2007-05-23 | 2007-10-10 | 重庆医科大学 | Image band parameter two-valued method and device using said method |
| CN101593276A (en) * | 2008-05-29 | 2009-12-02 | 汉王科技股份有限公司 | A kind of video OCR image-text separation method and system |
| CN101621615A (en) * | 2009-07-24 | 2010-01-06 | 南京邮电大学 | Self-adaptive background modeling and moving target detecting method |
Non-Patent Citations (1)
| Title |
|---|
| 楼晓燕，孙霞: "智能监控中图像预处理与去噪方法", 《嘉兴学院院报》, vol. 18, no. 3, 31 May 2006 (2006-05-31) * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108171235A (en) * | 2018-01-08 | 2018-06-15 | 北京奇艺世纪科技有限公司 | Title area detection method and system |
| CN108171235B (en) * | 2018-01-08 | 2021-01-22 | 北京奇艺世纪科技有限公司 | Title area detection method and system |
| WO2019137237A1 (en) * | 2018-01-09 | 2019-07-18 | 佛山科学技术学院 | Separate ocr recognition method and system thereof |
| CN108550099A (en) * | 2018-04-13 | 2018-09-18 | 北京焦点新干线信息技术有限公司 | The method and apparatus of watermark in a kind of removal image |
| CN108596916A (en) * | 2018-04-16 | 2018-09-28 | 深圳市联软科技股份有限公司 | Watermark recognition methods, system, terminal and medium similar in a kind of color |
| CN109214999A (en) * | 2018-09-21 | 2019-01-15 | 传线网络科技（上海）有限公司 | A kind of removing method and device of video caption |
| CN109214999B (en) * | 2018-09-21 | 2021-01-22 | 阿里巴巴（中国）有限公司 | Method and device for eliminating video subtitles |
| CN109903210A (en) * | 2019-01-04 | 2019-06-18 | 阿里巴巴集团控股有限公司 | Minimizing technology, device and the server of watermark |
| CN109903210B (en) * | 2019-01-04 | 2024-02-06 | 创新先进技术有限公司 | Watermark removal method, watermark removal device and server |
| CN111833231A (en) * | 2019-04-15 | 2020-10-27 | 阿里巴巴集团控股有限公司 | Watermark extraction method, device and system |
| CN113940090A (en) * | 2019-11-22 | 2022-01-14 | Inka 安特沃客有限公司 | Client-side evidence obtaining watermark device, system and method |
| CN113940090B (en) * | 2019-11-22 | 2023-08-01 | Inka 安特沃客有限公司 | Client evidence obtaining watermarking device, system and method |
| CN111798359A (en) * | 2020-05-19 | 2020-10-20 | 佛山市南海区广工大数控装备协同创新研究院 | Deep learning-based image watermark removing method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102567939A (en) | Separating method and device for visible watermark in image | |
| CN110678901B (en) | Information processing apparatus, information processing method, and computer-readable storage medium | |
| CN102456212A (en) | Separation method and system for visible watermark in numerical image | |
| US7532741B2 (en) | Data hiding in media | |
| CN103886545B (en) | Transformation domain printing-scanning resistant digital watermarking method based on computer-generated holography | |
| JP5015540B2 (en) | Digital watermark embedding device and detection device | |
| JP2008085695A (en) | Electronic watermark embedding apparatus and detection apparatus | |
| EP1526481A3 (en) | Object extraction based on color and visual texture | |
| CN104680473A (en) | Machine learning-based color image watermark embedding and detecting method | |
| CN102800043B (en) | The anti-counterfeiting information stacking method of printed matter, recognition methods and verifying attachment | |
| US9033371B2 (en) | Counterfeit prevention using miniature security marks | |
| Suresh et al. | Indian currency recognition and verification using image processing | |
| CN110533704B (en) | Method, device, equipment and medium for identifying and verifying ink label | |
| WO2008035401A1 (en) | Electronic watermark embedding device and detection device | |
| WO2001008098A8 (en) | Object extraction in images | |
| JP2006259885A (en) | Method of recognizing signboard and sign using color image | |
| JP2010136206A (en) | Image processing method and image processing apparatus | |
| CN106327416B (en) | A kind of site water mark method based on printed matter | |
| CN109859220B (en) | Linear image segmentation method | |
| CN101493927A (en) | Image reliability detecting method based on edge direction characteristic | |
| KR102180924B1 (en) | System and Method for Embedding and Extracting Digital Watermark Using QR Code | |
| CN104517259B (en) | Embedding algorithm method and apparatus in color text file | |
| CN108446701A (en) | A kind of best bounding volume method of writing brush word | |
| RU2431192C1 (en) | Method of inserting secret digital message into printed documents and extracting said message | |
| Kawaguchi | BPCS-steganography–principle and applications |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20120711 |